一种新的泊位飞机阴影分割方法

飞机泊位自动引导系统中,识别跟踪定位泊位飞机时,会出现灰度较低区域被误判为阴影区域的现象,严重影响识别效果,因此分割飞机阴影是泊位成功与否的关键技术。提出一种将形态学的幂变换与无边界主动轮廓线模型结合的阴影分割方法,首先利用多方向权重形态学对泊位飞机图像中的噪声和微小的干扰区域进行滤除,然后进行幂变换来增强图像质量,最后利用无边界主动轮廓线模型分割阴影。实验结果表明,该方法不仅能够滤除噪声,保留图像细节,而且能提高分割阴影的精确度,并且抑制了泊位飞机分割产生的＂拖尾＂现象,有效地检测出运动飞机的阴影。

New method of shadow segmentation about docking aircraft

During of aircraft auto-docking guide,identification and tracking aircraft positioning berth,lower gray area appears to be mistaken for the phenomenon of the shadow area,seriously affect the recognition results.Shadow segmentation of aircraft is one of the most key factors about succeeding in inducting aircraft docking.A new shadow segmentation method was proposed based on mathematics morphology power transform and active edgeless contour model syncretism.First using multi-direction weighted mathematics morphology filtering,noise and minute disturb area on docking aircraft image is filtered.Then processing power transform to buildup image quality,finally using active edgeless contour to segmentation shadow.Experimental results show that this method not only filters noise and preserves image details,improves the shadow segmentation precision,but also reduces the times of iteration.It also exploits the potential that active edgeless contour can be used in images with complex backgrounds.Consequently preferably solve docking aircraft image object partition question.